DocumentCode :
81219
Title :
Scene Text Recognition Using Structure-Guided Character Detection and Linguistic Knowledge
Author :
Cun-Zhao Shi ; Chun-Heng Wang ; Bai-Hua Xiao ; Song Gao ; Jin-Long Hu
Author_Institution :
State Key Lab. of Manage. & Control of Complex Syst, CASIA, Beijing, China
Volume :
24
Issue :
7
fYear :
2014
fDate :
Jul-14
Firstpage :
1235
Lastpage :
1250
Abstract :
Scene text recognition has inspired great interests from the computer vision community in recent years. In this paper, we propose a novel scene text-recognition method integrating structure-guided character detection and linguistic knowledge. We use part-based tree structure to model each category of characters so as to detect and recognize characters simultaneously. Since the character models make use of both the local appearance and global structure informations, the detection results are more reliable. For word recognition, we combine the detection scores and language model into the posterior probability of character sequence from the Bayesian decision view. The final word-recognition result is obtained by maximizing the character sequence posterior probability via Viterbi algorithm. Experimental results on a range of challenging public data sets (ICDAR 2003, ICDAR 2011, SVT) demonstrate that the proposed method achieves state-of-the-art performance both for character detection and word recognition.
Keywords :
Bayes methods; handwriting recognition; text analysis; Bayesian decision view; character sequence; computer vision community; global structure informations; integrating structure guided character detection; linguistic knowledge; part based tree structure; posterior probability; scene text recognition; structure guided character detection; word recognition; Character recognition; Feature extraction; Image color analysis; Image recognition; Object recognition; Optical character recognition software; Text recognition; Character recognition; Scene text recognition; character recognition; cropped word recognition; part-based tree-structured models; part-based tree-structured models (TSMs); posterior probability; scene text recognition; word spotting;
fLanguage :
English
Journal_Title :
Circuits and Systems for Video Technology, IEEE Transactions on
Publisher :
ieee
ISSN :
1051-8215
Type :
jour
DOI :
10.1109/TCSVT.2014.2302522
Filename :
6727574
Link To Document :
بازگشت